· Sophie · Behind the Build  · 2 min read

Behind the Build: How We Designed a Math App That Kids Actually Use

Most educational apps are abandoned within a week. We set out to build one that sticks — here's what we learned about motivation, design, and the science of habit.

Most educational apps are abandoned within a week. We set out to build one that sticks — here's what we learned about motivation, design, and the science of habit.

“Why would a kid open this app tomorrow?”

That single question drove every decision we made.

The Research That Changed Our Direction

We interviewed 30+ parents and observed 50+ kids using existing math apps. Three patterns emerged:

  1. Gamification wears off — stickers and stars stop mattering after day 3
  2. Kids sense “practice” apps — if it feels like homework, they avoid it
  3. Context matters most — kids engage when they see why the math matters

Our Design Principles

1. Math First, Game Second

Instead of wrapping math problems in a game (which feels hollow), we made the math itself interesting.

Concrete examples > abstract problems.

2. Streaks That Mean Something

We don’t count consecutive days. We count problems solved with understanding. If a child redoes a tough problem until they get it, that counts double.

3. Parent Dashboard, Not Parent Control

Parents can see what their child is learning — and more importantly, how they’re thinking about it. The dashboard shows:

  • Concepts mastered vs. in progress
  • Time spent per topic
  • Common mistake patterns

Not time limits, not lockouts. Trust over surveillance.

The Technical Stack

LayerTechnologyWhy
FrontendReact + TypeScriptType safety, huge ecosystem
BackendGo + PostgreSQLPerformance, reliability
ContentMDXFlexible, version-controlled lessons
AnalyticsCustom event pipelinePrivacy-first, no third-party tracking

What We’d Do Differently

If we started over tomorrow:

  1. Ship sooner — we spent too long perfecting the onboarding
  2. Test with real kids earlier — paper prototypes work; you don’t need code
  3. Fewer features, deeper ones — breadth is tempting; depth is what teaches

Building better learning tools, one iteration at a time. See what we’re working on now

Back to Blog

Related Posts

View All Posts »
The Stand-Up That Never Ended

The Stand-Up That Never Ended

Monday morning stand-up. 9am. Nobody had done anything on Sunday. Nobody had done anything before coffee. The stand-up became an existential debate about what "done" even means. Crystal pulled up charts. Manny gave a speech. Justin appeared for no reason.